Data processing system for estimating disease progression rates

Abstract

A method for treatment of a disease by monitoring a progression of the disease includes obtaining image data including a representation of diseased cells of a patient. Based on the type of the disease, one or more features to extract from the image data are determined, the features each representing a physical parameter of at least one of the diseased cells represented in the image data. A feature vector is formed from the extracted features. A machine learning model is selected, and the feature vector is processed using the machine learning model. The machine learning model is trained with labeled image data representing instances of diseased cells having the disease and associating scores representing predicted rates of disease progression with the respective instances of diseased cells having the type of disease. Based on the processing, a score is determined that represents a predicted rate of disease progression indicated by the image data.

What is claimed is: 
     
         1 . A method for treatment of a disease by monitoring a progression of the disease, the method comprising: obtaining image data including a representation of at least one disease cell of a patient, the at least one disease cell having a type of disease, the obtaining comprising: imaging, by an imaging device, the at least one disease cell at multiple instances over a time period to generate image data; for each instance of the multiple instances, extracting values from the image data for one or more features for the at least one disease cell, the one or more features each representing a physical parameter of at least one of at least one disease cell represented in the image data; inputting the values for the one or more features into a machine learning model, the machine learning model being trained with labeled image data that is associated with a same patient demographic as a patient demographic associated with the at least one disease cell; processing the values for the one or more features using the machine learning model, the machine learning model being trained with labeled image data representing instances of diseased cells having the type of disease and a given shape, the labeled image data associating scores representing predicted rates of disease progression with the respective instances of diseased cells having the type of disease and different shapes; determining, non-invasively and based on the processing, a score representing a predicted rate of disease progression indicated by the image data, the predicted rate of disease progression representing a comparison between a rate of progression of the disease in the at least one disease cell with a baseline progression from the labeled image data, the comparison being for a unique molecular profile and a unique cellular profile of the patient; and based on the score, generating a treatment regimen for how to perform a treatment including one or more of radiation therapy, chemotherapy, or tumor treatment fields (TTT) based on the predicted rate of disease progression; and performing the treatment based on the generated treatment regimen for how to perform a treatment. 
     
     
         2 . The method of  claim 1 , wherein cells of the image data are tagged with a protein tag, and wherein extracting the one or more features from the image data comprises identifying cells having the protein tag in the image data. 
     
     
         3 . The method of  claim 2 , wherein the protein tag is configured to fluorescently label cells in an NF-κB signal pathway, and wherein the type of disease comprises a glioblastoma multiforme (GBM) tumor, and wherein the method further comprises:
 including, with the one or more features values, based on the NF-κB signal pathway that is labeled, a feature value representing the NF-κB signal pathway; 
 processing, using the machine learning model, the feature value; and 
 determining, based on the processing, the score representing the predicted rate of disease progression, wherein the disease progression represents a growth rate of the GBM tumor. 
 
     
     
         4 . The method of  claim 1 , wherein the image data include a series of images including the disease cells of the patient and captured at time intervals, and wherein each image of the series of images is associated with a respective score to form a sequence of scores for the patient for each of the time intervals. 
     
     
         5 . The method of  claim 4 , wherein a length of one or more of the time intervals is automatically adjusted based on a determined growth rate of the diseased cells. 
     
     
         6 . The method of  claim 1 , wherein the type of disease includes at least one of a cancer, a bacteria, or a virus. 
     
     
         7 . The method of  claim 1 , wherein the disease cells are constituents of brain tissue, and wherein the one or more features include one or more of a cortical mean thickness value of the brain tissue, an inner-cortical surface area of the brain tissue, a mid-cortical surface area of the brain tissue, a pial-cortical surface area of the brain tissue, grey matter (GM) volume of the brain tissue, cerebrospinal fluid (CSF) volume of the brain tissue, white matter (WM) volume of the brain tissue, and a total volume for GM, CSF, and WM for each region of interest in the brain tissue. 
     
     
         8 . The method of  claim 1 , wherein the one or more features of include one or more of a size of a disease cell or group of disease cells, a shape of a disease cell or group of disease cells, a size of a nucleus of a disease cell, a shape of a nucleus of a disease cell, a cell density of the disease cells in tissue represented in the image data, and a pattern of cell migration of one or more of the disease cells relative to neighboring cells. 
     
     
         9 . The method of  claim 1 , wherein the machine learning model comprises a convolutional neural network (CNN), wherein one or more synapse weights are set based on the labeled image data. 
     
     
         10 . The method of  claim 1 , wherein the predicted rate of disease progression represents a rate of proliferation of diseased cells. 
     
     
         11 . A system for treatment of a disease by monitoring a progression of the disease, the system comprising: an imaging device configured to image at least one disease cell at multiple instances over a time period to generate image data; at least one processing device; and a memory in communication with the at least one processing device, the memory storing instructions that, when executed by the at least one processing device, cause the at least one processing device to perform operations comprising: obtaining the image data including a representation of at least one disease cell of a patient, the at least one disease cell having a type of disease; for each instance of the multiple instances, extracting values from the images data for one or more features for the at least one disease cell, the one or more features each representing a physical parameter of at least one of at least one disease cell represented in the image data; inputting the values for the one or more features into a machine learning model, the machine learning model being trained with labeled image data that is associated with a same patient demographic as a patient demographic associated with the at least one disease cell; processing the values for the one or more features using the machine learning model, the machine learning model being trained with labeled image data representing instances of diseased cells having the type of disease and a given shape, the labeled image data associating scores representing predicted rates of disease progression with the respective instances of diseased cells having the type of disease and different shapes; determining, non-invasively and based on the processing, a score representing a predicted rate of disease progression indicated by the image data, the predicted rate of disease progression representing a comparison between a rate of progression of the disease in the at least one disease cell with a baseline progression from the labeled image data, the comparison being for a unique molecular profile and a unique cellular profile of the patient; and based on the score, generating a treatment regimen for how to perform a treatment including one or more of radiation therapy, chemotherapy, or tumor treatment fields (TTT) based on the predicted rate of disease progression; and causing performance of the treatment based on the generated treatment regimen for how to perform a treatment. 
     
     
         12 . The system of  claim 11 , wherein cells of the image data are tagged with a protein tag, and wherein extracting the one or more features from the image data comprises identifying cells having the protein tag in the image data. 
     
     
         13 . The system of  claim 12 , wherein the protein tag is configured to fluorescently label cells in an NF-κB signal pathway, and wherein the type of disease comprises a glioblastoma multiforme (GBM) tumor, and wherein the operations further comprise:
 including, with the one or more features values, based on the NF-κB signal pathway that is labeled, a feature value representing the NF-κB signal pathway; 
 processing, using the machine learning model, the feature value; and 
 determining, based on the processing, the score representing the predicted rate of disease progression, wherein the disease progression represents a growth rate of the GBM tumor. 
 
     
     
         14 . The system of  claim 11 , wherein the image data include a series of images including the disease cells of the patient and captured at time intervals, and wherein each image of the series of images is associated with a respective score to form a sequence of scores for the patient for each of the time intervals. 
     
     
         15 . The system of  claim 14 , wherein a length of one or more of the time intervals is automatically adjusted based on a determined growth rate of the diseased cells. 
     
     
         16 . The system of  claim 11 , wherein the type of disease includes at least one of a cancer, a bacteria, or a virus. 
     
     
         17 . The system of  claim 11 , wherein the disease cells are constituents of brain tissue, and wherein the one or more features include one or more of a cortical mean thickness value of the brain tissue, an inner-cortical surface area of the brain tissue, a mid-cortical surface area of the brain tissue, a pial-cortical surface area of the brain tissue, grey matter (GM) volume of the brain tissue, cerebrospinal fluid (CSF) volume of the brain tissue, white matter (WM) volume of the brain tissue, and a total volume for GM, CSF, and WM for each region of interest in the brain tissue. 
     
     
         18 . The system of  claim 11 , wherein the one or more features of include one or more of a size of a disease cell or group of disease cells, a shape of a disease cell or group of disease cells, a size of a nucleus of a disease cell, a shape of a nucleus of a disease cell, a cell density of the disease cells in tissue represented in the image data, and a pattern of cell migration of one or more of the disease cells relative to neighboring cells. 
     
     
         19 . One or more non-transitory computer readable media storing executable instructions that, when executed by at least one processing device, cause the at least one processing device to perform operations for treatment of a disease by monitoring a progression of the disease, the operations comprising: obtaining image data including a representation of at least one disease cell of a patient, the at least one disease cell having a type of disease, the obtaining comprising: imaging, by an imaging device, the at least one disease cell at multiple instances over a time period to generate image data; for each instance of the multiple instances, extracting values from the images data for one or more features for the at least one disease cell, the one or more features each representing a physical parameter of at least one of at least one disease cell represented in the image data; inputting the values for the one or more features into a machine learning model, the machine learning model being trained with labeled image data that is associated with a same patient demographic as a patient demographic associated with the at least one disease cell; processing the values for the one or more features using the machine learning model, the machine learning model being trained with labeled image data representing instances of diseased cells having the type of disease and a given shape, the labeled image data associating scores representing predicted rates of disease progression with the respective instances of diseased cells having the type of disease and different shapes; determining, non-invasively and based on the processing, a score representing a predicted rate of disease progression indicated by the image data, the predicted rate of disease progression representing a comparison between a rate of progression of the disease in the at least one disease cell with a baseline progression from the labeled image data, the comparison being for a unique molecular profile and a unique cellular profile of the patient; and based on the score, generating a treatment regimen for how to perform a treatment including one or more of radiation therapy, chemotherapy, or tumor treatment fields (TTT) based on the predicted rate of disease progression unique molecular profile and the unique cellular profile of the patient; and causing performance of the treatment based on the generated treatment regimen for how to perform a treatment. 
     
     
         20 . The method of  claim 1 , further comprising:
 tagging the at least one disease cell with a cell tag in a cell cultivation environment;   obtaining the image data of the at least one disease cell at the multiple instances over the time period;   for each instance of the multiple instances, extracting the values for the one or more features for the at least one disease cell, the one or more features comprising a size of the at least one disease cell, a surface smoothness factor for the at least one disease cell, and a cell count associated with the at least one disease cell;   inputting the values for the one or more features into the machine learning model; and   generating, based on the inputting, the score representing the predicted rate of disease progression.   
     
     
         21 . The method of  claim 1 , further comprising:
 training the machine learning model with the labeled image data representing instances of diseased cells having the type of disease, the labeled image data associating the scores representing the predicted rates of disease progression with the respective instances of diseased cells having the type of disease.